Suppose an asset costs $100 in the cash market and a derivative on it trades at $102. Buying the asset and selling the derivative appears to lock in $2. The interesting part is what has to happen before that $2 becomes cash you can keep.
Direction and the spread
Buy one unit at $100 and short one equivalent unit at $102. Later, close both at $110. The spot position earns $10 and the short loses $8, for a combined $2 before costs. If both instead finish at $90, the spot loses $10 and the short earns $12. The same $2 remains.
The trade is exposed to the difference between the two prices, called the basis. If you close when spot is $110 but the derivative is $115, the combined result is $10 minus $13: a $3 loss. You hedged the common movement, but the spread widened against you.
Expiry and funding create different trades
A dated future has a specified settlement process. If its final payoff corresponds to the asset you own, holding through settlement can provide a route to convergence. Financing, delivery, commissions, and differences in the settlement reference still belong in the calculation.
A perpetual has no comparable expiry date. Funding encourages its price to remain near the reference, but it does not guarantee that an entry premium disappears by your preferred exit. On Hyperliquid, funding is exchanged between longs and shorts and depends partly on the premium; the documented payment calculation uses position size and the oracle price. See the funding specification.
For a long-spot, short-perp position, positive funding paid by longs is income. Negative funding is a cost. Keep the basis change and the funding payments as separate lines: one is the change in the gap between prices; the other is cash paid while the position is open. Counting the entry premium as income immediately and then adding all subsequent funding overstates what has been earned.
Why doesn't everyone collect the spread?
Someone must finance the asset, supply collateral to the short, and remain able to move money when prices change. Those requirements become harder precisely when many traders want leverage.
A BIS study of crypto carry links large spreads to demand for leveraged exposure and limits on arbitrage capital. That historical finding offers a useful interpretation: a high advertised carry rate can be compensation for providing scarce balance sheet and bearing stress, rather than evidence of an overlooked free return.
Imagine the asset doubles. The spot gain approximately offsets the short's loss economically, but the short venue may demand cash now. The matching gain may sit in an asset held elsewhere. Transfers can take time, withdrawals can stop, and collateral may receive a haircut. A portfolio can be hedged on paper and still be unable to meet the next margin requirement.
Return on committed capital
Suppose you commit $10,000 to spot, $2,000 to derivative margin, and $1,000 to a reserve. Over a month, funding and basis changes produce $200; financing and execution consume $80. The $120 net profit is about 0.92% of the $13,000 committed, not 6% of the $2,000 margin deposit.
The reserve is not wasted money. It is part of what permits the strategy to survive. Annualizing one unusually profitable month also assumes an opportunity and a funding rate that may not persist.
Closing both legs
Before entering, estimate the cost of closing both legs at the size you intend to trade. Record the funding rate at which the position stops being attractive and the maximum temporary margin demand you can accommodate. If one leg becomes inaccessible, decide which exposure you can close first without creating an unplanned directional bet.
The apparent $2 spread at entry is only one part of the return. It becomes a profitable trade if convergence and funding exceed financing and execution costs, and the account can meet the intervening demands for collateral.